Monocular Vision Based Object Recognition and Tracking for Intelligent Robot.

The intelligent robot is a system that can detect the surroundings and make corresponding actions according to environment changes. It can perform the corresponding functions according to the pre-set instructions. Navigation is one of the important directions for robot research. The robot can move from the starting point to the destination under the control of corresponding module. The main work of this paper is based on the visual tracking of markers and general targets. For tracking of markers, the EPnP algorithm is used to solve the camera's position and orientation relative to marker so that the robot can track it. For general target tracking, the AdaBoost algorithm is employed to recognize the target. Then we can get the relative position and realize tracking by using the RGBD information.